Experimental Data Sets for the study "Benchmarking a $(\mu+\lambda)$ Genetic Algorithm with Configurable Crossover Probability"
DOI10.5281/zenodo.3753086Zenodo3753086MaRDI QIDQ6690420FDOQ6690420
Dataset published at Zenodo repository.
Thomas Bäck, Hao Wang, Furong Ye, Carola Doerr
Publication date: 14 April 2020
Copyright license: Creative Commons Attribution 4.0 International
This is the experimental result of the study Benchmarking a (+) Genetic Algorithm with Configurable Crossover Probability. A novel(+) GA is proposed and benchmarked, in which we stochastically determine whether to apply the crossover operator either for each individual or generation with a crossover probability\(p_c\).This data setconsistsof two parts: The results of (+) GA on 25 pseudo-Boolean problems defined in IOHprofiler (https://iohprofiler.github.io/) with the followingsetup:\(\mu \in \{10, 50, 100\}, \lambda \in \{1, \lceil\mu/2\rceil, \mu\}, p_c\in\{0, 0.5\}.\) IOHprofiler_Problems_standard_bit_mutation.csv -- the (+) GA with standard bit mutation. IOHprofiler_Problems_fast_mutation.csv -- the (+) GA with fastmutation. The results of (+) GA on OneMax and LeadingOnes problemswith the following setup:\(n \in \{64,100,150,200,250,500\}, \mu \in \{2,3,5,8,10,20,30,...,100\}, \\ \lambda \in \{1, \lceil \mu/2 \rceil, \mu\}, \text{and }p_c \in \{0.1 k \mid k \in [0..9]\}\cup\{0.95\}.\) OneMax_raw.csv -- the fixed-target running time/first hitting time from 100 independent runs for target values in\([1..n]\). OneMax_summary.csv -- the mean, median, standard deviation, some quantiles, expected running time (ERT), the number of successful runs, and the success ratefrom 100 independent runs for target values in\([1..n]\). LeadingOnes_raw.csv -- the same with OneMax_raw.csv for LeadingOnes. LeadingOnes_summary.csv -- the same with OneMax_summary.csv for LeadingOnes. Contact: if you have any questions or suggestions, please feel free to contactFurong Ye or Carola Doerr.
This page was built for dataset: Experimental Data Sets for the study "Benchmarking a $(\mu+\lambda)$ Genetic Algorithm with Configurable Crossover Probability"
